Several converging lines of evidence point towards the role of prenatal stress and anxiety in the aetiology of autism spectrum disorders , although population based human studies are scarce.1 One large Danish study investigating this issue used bereavement as a proxy for stress during pregnancy and observed associations which were lost in adjusted analysis (this included possible mediating factors that should arguably not be adjusted for). Two clinic based studies reported a significant effect of exposure to stressful life events on the diagnosis of autistic disorder in the child. Both these were small with retrospective data collection and may be open to selection and recall bias.2,3 There is also evidence from an ecological study describing higher incidence of autism in areas affected by tropical storms. Affective disorders in parents have also been reported to be 'independent' risk factors for autism in one study.4 However, reverse causality may explain this association since no study has ascertained parental diagnoses prior to the birth of the child, and caring for a disabled child may predispose parents to develop mood and anxiety disorders.

It is also not known if observed associations are largely due to a) genetic or biological factors related to stress and anxiety, b) foetal exposure to antidepressant or similar medication, c) due to early life environmental disadvantage associated with growing up with a stressed or mentally unwell parent, or d) chance findings as they have yet to be replicated in other well designed population based studies.

The ALSPAC data allows us to investigate these questions in some detail. First, stress and anxiety are measured in a number of ways, at different time points including adverse life events (possibly other measures of adversity- the 'stressors') as well as self reported anxiety (that may depict the individuals' response to stress, and may be used to differentiate between acute or chronic/ state or trait anxiety based on the number of times the woman appears stressed on questionnaires) along with possible confounders in this relationship including drug use and important socio-demographic characteristics.

1 Kinney DK, Munir KM, Crowley DJ, Miller AM. Prenatal stress and risk for autism. Neurosci Biobehav Rev 2008;32:1519-32.

2 Beversdorf DQ, Manning SE, Hillier A, Anderson SL, Nordgren RE, Walters SE, et al. Timing of prenatal stressors and autism. J Autism Dev Disord 2005;35:471-8.

3 Ward AJ. A comparison and analysis of the presence of family problems during pregnancy of mothers of "autistic" children and mothers of normal children. Child Psychiatry Hum Dev 1990;20:279-88.

4 Larsson HJ, Eaton WW, Madsen KM, Vestergaard M, Olesen AV, Agerbo E, et al. Risk factors for autism: perinatal factors, parental psychiatric history, and socioeconomic status. Am J Epidemiol 2005;161:916-25.